Center for Systems Integration and Sustainability, Michigan State University

Lit Review Headaches

Lit Review Headaches

Lit Review Headaches

Lit Review Headaches

Bibliometrix

Bibliometrix

<www.bibliometrix.org>

Where do I start?

Export from Web of Science in Bibtex format!

Where do I start?

Export from Web of Science in Bibtex format!

Get 'R started!

Load packages…

  • The most important one:
library("bibliometrix") # for biblio-analytics
  • Other cool ones:
library("tidyr") # for table manipulations
library("dplyr") # for table manipulations
library("plyr") # for table manipulations
library("tidyverse") # for graphics/table management
library("ggplot2") # for graphics
library("ggmap") # for graphics
library("RColorBrewer") # for graphics
library("tm") # text analysis for word clouds
library("SnowballC") # text analysis for word clouds
library("wordcloud") # text analysis for word clouds

Get 'R started!

Load bibtex file

bib.dir <- paste0("D:\\Vero\\Documents\\R\\R_ladies_presentations\\bibliometrix\\")
niche.bib <- readFiles(paste0(bib.dir,"niche_theory_484.bib"))

Convert

niche.df <- convert2df(niche.bib,dbsource = "isi", format = "bibtex")

Quick biblioanalysis

niche.biblio <- biblioAnalysis(niche.df, sep = ";")

Get summary

summary(niche.biblio, k=5, pause=FALSE) #k=5 is first 5 on list

Get 'R started!

Get summary

summary(niche.biblio, k=5, pause=FALSE) #k=5 is first 5 on list
  • Main information
  • Number of articles per year
  • Annual percentage growth rate
  • Most productive authors, countries
  • Most cited articles
  • Number of citations per country
  • Most relevant journals (most articles)
  • Most relevant keywords

Get 'R started!

Graphics

plot(x=niche.biblio, k=5, pause=FALSE) #k=5 is first 5 on list

Get 'R going: ggplot2!

  • Number of articles per year
art.yrs <- ddply(niche.df, .(PY), summarize, count=length(PY))

Get 'R going: tm/wordcloud!

  • Word cloud (title, abstract and keywords)
wordcloud(words=keyterms.df$word, freq=keyterms.df$freq, 
         min.freq=10, max.words=100, random.order=FALSE,
         rot.per=0.0, colors=brewer.pal(8, "Paired"))

Get 'R going: tm/wordcloud!

  • Word cloud (title, abstract and keywords)

Get 'R going: tm/wordcloud!

  • Investigate your keywords

Get 'R going: tm/wordcloud!

  • Investigate your keywords
keyterms.df[grep("climate", keyterms.df$word),]
##                      word freq
## climate           climate  156
## climates         climates    9
## macroclimate macroclimate    3
## paleoclimate paleoclimate    1

Get 'R going: tm/wordcloud!

  • Find word associations
options(width=65) # ensure width
findAssocs(term.mtrx1, terms = "climate", corlimit = 0.35)
## $climate
##        change vulnerability  phenological       douglas 
##          0.68          0.55          0.50          0.46 
##         frost      interior      nitschke          taca 
##          0.46          0.46          0.46          0.46 
##      columbia        future       respond     phenology 
##          0.41          0.38          0.38          0.38 
##       merging           pau   photoperiod      reponses 
##          0.38          0.38          0.38          0.38 
##       british      climatic 
##          0.37          0.35

Reeeally get 'R going: ggmap!

  • Number of articles per country (first author affiliation)

Thank you!

Questions?

 

 

Contact:

  • verofrans[at]gmail.com

  • twitter: @VFFRANS